Skip to main content
Glama

marginalia

A Claude Code skill: render any Markdown into a browser-based, click-to-comment thread. Comments return to Claude as clean MCP tool_results; replies render as in-page cards that are themselves annotatable (threaded).

Install

As a Claude Code plugin (recommended):

/plugin marketplace add Jin-HoMLee/marginalia
/plugin install marginalia@marginalia
# restart Claude Code

The bundled .mcp.json registers the MCP server automatically. If Python deps are missing, run python3 -m pip install -r requirements.txt.

Manual (skill + MCP at user scope, no plugin):

python3 -m pip install -r requirements.txt
./scripts/register.sh        # registers the MCP server at user scope
# restart Claude Code

Related MCP server: nia-link

Test

python3 -m pytest -v

Architecture

One process, two faces: a FastMCP stdio server (the tools Claude calls) and a localhost HTTP server (serves the page, receives comments). They share one in-process ThreadStore. See docs/superpowers/specs/2026-06-16-marginalia-mcp-skill-design.md for the full design and docs/superpowers/plans/2026-06-16-marginalia-mcp-skill.md for the build plan.

A
license - permissive license
-
quality - not tested
C
maintenance

Maintenance

Maintainers
Response time
Release cycle
Releases (12mo)
Commit activity

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/Jin-HoMLee/marginalia'

If you have feedback or need assistance with the MCP directory API, please join our Discord server